PLINK: Identical A1 and A2 alleles on line 215

Question

[ommegang]

related: How to bypass SNPs with identical A1 and A2 alleles in PLINK?

I have a huge dataset in bim/bed/fam format, for many individuals and separated by chromosome. I'm trying to merge the datasets into a single file using Plink 1.9. I'm getting the error message

Error: Identical A1 and A2 alleles on line 215 of imp_chr1.bim.

The bim file line is

1 rs3131984 0 715348 G G

The data is imputed, and I'd like to keep those alleles. Is there any way to do this?

I've also tried to remove them by creating a list and using --exclude, but I'm getting the same error.

[laurentelkrief9]

Hi did you figure this out ? I think I am working on the same dataset and cannot figure out how to bypass this error

Thanks

Answer 1

[chrchang523]

This is a malformed .bim. You should figure out the original cause of the problem and fix that, if possible.

Failing that, can you post the .log file of your failed --exclude attempt? I just verified that the current plink 1.9 and 2.0 builds allow you to exclude variants like this.

[ommegang]

Are you sure this is malformed? I thought this could happen when doing imputation?

[chrchang523]

That's a bug in the imputation tool. The correct .bim would have one of the alleles set to '0'.